Hyperpolarized xenon-129 (HXe) MRI has been demonstrated to have exquisite sensitivity for functional diagnostic imaging of lungs, likely the most technically promising and commercially viable technique. Full characterization of pulmonary health may benefit from as many as eight different multi-liter protocols, which should ideally be completed in well under an hour. Three years ago our group identified an efficient method for polarization and developed an apparatus to exploit the high flow, low pressure polarization regime, increasing output by more than a factor of ten to 1 L/hr at 50%. In our Phase 1 proposal we argued that assembling a polarizer with several heat-conducting channels operating in parallel would permit a scale up by a factor of the number of channels, assuming similar conditions of temperature, pressure, laser illumination, and cell geometry. Furthermore, if the laser spectrum could be narrowed, then dropping the pressure by a factor of two could yield another two-fold improvement in production rate. During Phase 1 we developed new fabrication techniques, assembled a copper prototype, adapted our existing laser and magnetic infrastructure for its installation, and characterized its performance. We confirmed predictions, that the multi-channel polarizer column delivers HXe with >50% polarization at a rate of two liters per hour per hundred watts of narrowed laser power, up to 250 watts, a six-fold improvement over our previous world's best production rate. We discovered that birefringence arising from thermal stresses in a few laser elements was compromising polarization demonstrations of our copper-column technology up to the full kilowatt of available laser power, a problem that can now be easily corrected. In Phase 2 we propose full characterization of our new prototype polarizer, demonstrating twenty- fold scale up of xenon polarizer production output. We will characterize its performance over wide ranges of gas mixture, pressure, temperature, flow rates, and laser power. We will probe the low pressure limit, including a more complete characterization of the nitrogen partial pressure required for quenching. We will develop technology to monitor remaining rubidium and to sequester condensed rubidium, extending the service lifetime. We will incorporate technology to recirculate the pure buffer gases, minimizing their consumption. We will assemble these technologies into a polarizer platform with expanded gas flow and freeze-out capabilities, and a more versatile and efficient xenon delivery system. We will optimize a prototype high-flow polarizer capable of producing MagniXene} at 20 L/hr at 50% polarization, which meets the needs of clinical partners and pharmaceutical customers. After this Phase 2 project the polarizer technology will be completed, and ready for the detailed engineering and extensive documentation required for FDA approval as a Class 2 device. [unreadable] [unreadable] [unreadable]